| 摘要 |
<p>PURPOSE: A method and an apparatus for compressing images are provided to easily compress an image into a predetermined file size by performing grouping and iterative estimation of member entities of the image based on image features, without processing the entire image. CONSTITUTION: An apparatus for compressing an image A method of compressing an image comprises the steps of: grouping member of the image entities into a plurality of groups based on each image features (1002); selecting at least one group representative from at least one of the plurality of groups (1010); estimating final control parameters for each of the group representatives in an iterative manner (1011); and compressing the image based on the estimated control parameters. [Reference numerals] (1001) Generating all image features on all MCU; (1002) Define groups; (1003) MCU counter start : i=0; (1004) Group counter start : m=0; (1005) Grp_m matches MCU_i feature values; (1006) Intensity of Grp_m increases; (1009) Calculate quantization tables on QF_MIN; (1010) Selecting at least one group representative from at least one of the plurality of groups; (1011) Estimating final control parameters for each of the group representatives in an iterative manner; (1012) EST_MIN <= target size?; (1013) Calculate initial QF_0; (1014) State=failure; (AA) Start; (BB,EE,FF,GG,II,KK) No; (CC,DD,HH,JJ) Yes; (LL) End</p> |
